D O I M EAN W HAT I S AY AND S AY W HAT I MEAN ? A C ROSS - C ULTURAL A PPROACH TO THE U SE OF E MOTICONS & E MO- JIS IN CMC M ESSAGES
¿Digo lo que siento y siento lo que digo? Una aproximación transcultural al uso de los emoticonos y emojis en los mensajes en CMC
Dra. Lifen CHENG
Profesora titular de la Universidad de Salamanca (España) E-mail: [email protected]
Fecha de recepción del artículo: 01/06/2017 Fecha de aceptación definitiva: 23/10/2017
ABSTRACT
The use of emoticons and emojis among online messaging users has achieved a globalized level. This article aims at examining how certain emojis and emoticons are chosen by message senders to regulate the intensity of emotions in their messages. Or, whether by means of using them, message senders wish to blur textual rigidity and show interac- tion empathy during the communication process. This study analyzes the use of these symbols and graphic elements in message exchanges through CMC and SNS, where users from different culture backgrounds convey their feelings.
To do so, two empirical studies with a methodological design were carried out: One pilot study was intended as re- search instrument validation. The second was a survey study. By testing online messaging habits, this study enabled the researchers to observe whether the senders were inclined to using linguistic text use only, or whether they pre- ferred including an additional emoticon or emoji in their message to communicate their feelings. Results of this re- search show that, to a certain extent, message senders’ different cultural backgrounds could influence their emotion- al expressivity level.
Key words: CMC, culture variability, emojis, emoticons, SNS, ICT.
RESUMEN
El uso de los emojis y emoticonos entre los usuarios de mensajería online es un fenómeno global y generalizado. El presente estudio analiza las preferencias entre los usuarios de unos emojis y emoticonos sobre otros con el fin de adecuar mejor la expresión de sus mensajes con la emoción que contienen. Lo hacen así para intensificar o difuminar la rigidez textual. También, para manifestar su propia empatía durante las interacciones del proceso comunicativo.
Este trabajo analiza el uso de estos símbolos y elementos gráficos en el intercambio de mensajes online por medios sociales; a través de ellos los usuarios, según su diferente procedencia cultural, van a trasmitir aquellos estados emo- cionales que más se sientan afectados. Para ello, se han llevado a cabo estudios empíricos basados en un diseño metodológico que consta de dos fases de investigación: Un estudio piloto para la validación del instrumento inves- tigador y un segundo estudio basado en encuesta. Los resultados observados indican que la procedencia cultural podría influir, hasta cierto punto, en el grado de expresividad de su gente a la hora de elegir entre 1) el uso de texto verbal único, 2) los elementos de signo en el teclado, 3) la inclusión de un emoticono o, 4) un emoji facial expresi- vo, cuando utilizan la mensajería online de medios y redes sociales para comunicar sus sentimientos o emociones.
Palabras clave: CMC, emojis, emoticonos, SNS, TICs, variabilidad cultural.
1. I
NTRODUCTIONThe use of applications derived from information and communication technologies (ICT) has be- come part of the everyday life of many people in our planet. Over the last decade, access to the inter- net has been adopted by the general population as a social vehicle for their interpersonal interactions.
CMC, which stands for Computer-mediated communication, is an umbrella term that includes any human communication that occurs through the internet and with ICT, and whose development and expan- sion are constant all over the world and among all social levels.
Essentially, CMC is defined as any human communication that occurs through the use of two or more electronic devices. While the term has traditionally referred to those communications that occur via computer-mediated formats, it has also been applied to other forms of text-based interaction such as text messaging. Until recently, most research on CMC was focused on studies about the social effects of the different communication technologies with computer support (Adrianson, 2001), whereas many of the most recent analyses study social networks with social software support. In this regard, CMC is considered a process where there is an interaction of human data through one of more telecommunication systems and software connected online. These systems and software include:
email, Internet Relay Chat (IRC), instant messaging (IM), Usenet (Users Network), LISTSERV (mail- ing list servers —mailing lists that provide internet services based on emails and that include a group of electronic addresses used to send messages or announcements to the members of the list—), as well as SMS, acronym of Short Message Service, a text messaging system for mobile phones, as well as SNS (Social Networking Service).
CMC can take place synchronously and asynchronously. In the first mode, all the participants take part in the communication simultaneously. In the second one, there are time restrictions between the moment in which the message is sent and the moment in which the response is received, as in the case of emails. The key characteristics of CMC make it possible to record and retake the conversation, and they include formal communication and/or the anonymity of the user, depending on the type of software used —for example, IM (instant messaging)—. However, the emotions underlying a state- ment from a user may be difficult to interpret due to the lack of face-to-face communication. The lack of visual contact can generate misunderstandings in the form of unwanted negative effects. Therefore, emoticons represent a useful tool to bridge that gap.
It seems that this aspect has been partially overcome with the possibility to include emoticons and emojis in digital messages, although the consequences of their continued use are not very clear: Are users changing the way they think? Have our brains changed over the last few years with the appear- ance of emoticons and emojis? Are speaking patterns changing? Both emoticons and emojis are rec- ognized and processed by the brain as non-verbal information, which means that we “read” them as part of emotional communication. And we can ask yet one more question: Does their use vary across different cultures?
This last question is the main focus of our study, which analyzes whether there is a differentiated preference in the use of emoticons, emojis or verbal text alone among the speakers of different cul- tures when expressing their emotions or their empathy.
2. L
ITERATURER
EVIEWThe origin of emoticons and emojis
Although both terms are used interchangeably and often as synonyms, the truth is that they are different things. The term “emoticon” is a portmanteau of “emotion” and “icon”, and it refers to
emotion and images. They are commonly used in CMC, in which the typical factors of face-to-face communication are eliminated, such as the rhythm of enunciation, gestures, body distance, immediate feedback from the interlocutor, etc. (Dresner & Herring, 2010). Social presence and communication theories postulate that for communication to take place effectively, participants need to have a way to obtain immediate feedback or use natural language (Ekman, 1993). In general terms, CMC lacks the available channels and key elements for effective communication. This is due to the lack of a social context applied as a theoretical reference. Some researchers claim that CMC users have developed new non-verbal signs, such as paralinguistic expressions, which are called “emotex” or “emoticons” to overcome those limitations and, in fact, they are used by many users and netsurfers in their communi- cations over the internet. Ferris (1996: 34) defines “emotex” as non-verbal signals to emphasize emo- tion and suggests that an emotex consists of lexical replacements for non-verbal signs. In this regard, as the term implies, emoticons represent visual substitutes with the characteristics of a text which contain symbols to express different types of emotion. Their invention is attributed to Scott Fahlman, from the Carnegie Mellon University in the USA, who represented a smiling face with keyboard signs on the computer and was the first person to use these symbols three decades ago (Ptaszynski & Arak, 2010). These authors explain that Fahlman created the smiley symbol with three keyboard signs :-) on his computer, and another symbol, the frowny, with these signs :-(. Afterwards, he added another sym- bol “›” to the frowny for emphasis, thus creating the symbol angry ›:-(, to express the feeling of anger.
Originally, these combinations of keyboard signs were used to identify jokes in a scientific discus- sion forum about computing. Afterwards, their use was adapted to optimize computer-mediated inter- actions. As Falhlman explains on his website (https://www.cs.cmu.edu/~sef/sefSmiley.htm), he came up with the idea in 1982 as a response to the interpretation problems regarding the messages ex- changed on the electronic boards in which different topics were discussed online between teachers and students, who thought that it would be a good idea to explicitly mark posts that were not to be taken seriously because in a text-based online context, users did not have the additional information provided by body language or the tone of voice as in a face-to-face conversation, in which those body or aural nuances can provide information about the information or the tone of the message. In view of this situation, Fahlman thought that the sequence of characters :-) could be an elegant solution, because all the computers available back then, based on ASCII code, were equipped with these signs.
He also thought that they could be a complementary way of protecting other people’s feelings and express different states of mind (Ptaszynski & Arak, 2010). Since then, these symbols have become popular for thousands of people who use them on a daily basis, together with other emoticons that were created later, to express their mood, their agreement or disagreement, their state of mind, etc.
Emojis have later on represented an extraordinary creative advance. These are single, closed icons that portray graphically a wide range of emotions that can be interpreted visually. They are used to express the emotional states of each text strictly based on the symbols themselves, just as the use of non-verbal language is used in face-to-face communication. Emojis also help to emphasize or high- light the tone or the meaning of a message when it is being written (Crystle, 2001). Apart from making it easier to communicate a mood (Constantin & Kalyanaraman, 2002), emojis clarify textual messages through the screens, just as non-verbal language does in face-to-face communication (Walther &
D’Addario, 2001).
Emojis have evolved to become a new language. Their different designs provide new codes that stimulate their increased use by CMC users. Emoji designers are constantly creating new and exclusive characters to increase the range of feelings that can be conveyed, with greater detail and in an attempt to achieve a more effective understanding. Everyday new emoji models appear in the messaging sys- tems of Twitter, Facebook and mobile phones. Many new possibilities to share different emojis can be found in the chat and personal page messages in Facebook.
Although emoticons and the new emojis, which were created later, are very popular in the area of communication, they are not always compatible with all the devices, and in some cases they are shown with a different design. For this reason, there are different emojis depending on the cultural area cov- ered by each mobile device and its potential users.
Background studies about the use of emoticons and emojis in CMC
Previous studies claimed that CMC lacks any sign of non-verbal interaction and, as a conse- quence, internet users have adopted the use of emoticons to improve their communication. It is thought that these features will replace the signs lost in the discourses written on a computer. A re- view of the literature reveals that emoticons are used as socio-emotional providers in the context of CMC, where it has been shown that the use of emoticons and emojis has evolved to the point in which it has become a recurring element in almost all computer-mediated interaction modalities (Lo, 2008). Different researchers have studied them from different perspectives. For example, the psycho- logical approach tends to focus on emotions and on the role played by emoticons/emojis on the un- derstanding of messages. Linguists have analyzed emoticons with regard to speech, as a parameter in writing and in the search for a semantic and syntactic explanation, together with their functional classi- fication and visual aspects. Rather than merely being an internet para-language, the graphemes in emoticons have been compared to vocabulary morphemes with different functions regarding their meaning (Pierozak in Amaghlobeli, 2012). This would imply that emoticons can become units like morphemes, in which there may be derivation, inflection or abbreviation, but which would not be isolated units. A pedagogical study has even proposed the creation of an emoticon-based universal language which could be introduced into the education system (Jibril and Abrudllah, 2013).
In another study, Tossell et al. (2012) focused on understanding how emoticons were used in text messages. More specifically, they analyzed the differences in the variety of emoticons and the frequen- cy of use between men and women who send messages over social networks. The differences ob- served reveal that women send a higher number of messages with emoticons, whereas men use a more varied range of these elements. Another analysis shows that men use emoticons to create humor and show emotions whereas women use them to communicate sarcasm (Lee, 2003). On the other hand, Huffaker and Calvert (2006) analyzed the contents of different blogs and observed that blogs written by men had a higher number of emoticons than those written by women. Other studies suggest that women tend to use more a non-verbal communicative style in F2F encounters (friend-to-friend).
Derks et al. (2007) wanted to verify whether this also happened in CMC, and developed studies that focused on the differences between both sexes, with mixed results. On the internet, men did not use many emoticons when commenting on sports news where most of the readers were also men. How- ever, when men participate in mixed groups, they used emoticons more frequently. The authors sug- gest that both men and women want to clarify their emotional state through a reading of the contents.
Also, although the frequency of emoticons found in mixed forums were more or less equivalent, each sex used them in a different way. This finding seems to indicate that in all the methods used to analyze CMC exchanges, there are no systematic differences between both genders because the emoticons are used differently depending on the tasks, contexts and media that are being used. The same thing hap- pens with facial expressions and other non-verbal communication types, which are recurrent to com- municate social findings and emotions; and to make the meaning of messages clearer (Derks et al., 2007).
One important finding observed by the researchers is the lack of contextual information in CMC exchanges, which affects the comprehension of contexts and the tone of the messages. This lack of context has been considered by the emitters of the messages as the culprit of creating the perception of electronic messages as being unpleasant or offensive (Jenson, 2005; Derks et al., 2007). This proves that emoticons can supply this information in order to improve understanding in CMC. Emoticons
are defined as the graphic representation of facial expressions which are embedded in electronic mes- sages. Derks et al. (2007) have also observed that emoticons are more commonly used in synchronous communication. This suggests that the immediacy of their use can promote a better communicative relationship. Other empirical studies observed that the online visualization of electronic texts without emoticons led to a misinterpretation of the intended tone of the message and the attitude of the send- ers (Lo, 2008). The inclusion of emoticons helped readers to understand better the level and direction of the emotional context around the message sent over the internet. In instant messaging, positive emoticons were used more often than negative ones, and their use intensifies the validity of the mes- sage. In that study, emoticons were mainly identified as a way to express emotion, strengthen the messages and show humor or sarcasm.
Rezabek and Cochenour (1998) analyzed a set of emails with the program LISTSERV to deci- pher the content of emoticons and their frequency of use. These researchers observed that 1-25% of the emails tin their study contained emoticons, and that the LISTSERV entry with the highest num- ber of analyzed messages only contained 6% of all emoticons. Afterwards, Ling (2005) examined 882 messages through telephone interviews to collect data and found that only 6% of those messages contained emoticons. The authors claim that there are many factors that may have had an influence on such a significant variability between different messages, such as the strength of the social bond, gender, age, or geographic distribution, which were not assessed in the study. Due to the brevity of communications through SMS (Short Message Service) and the fact that it is used both synchronously and asynchronously (Kasesniemi & Rautianen, 2002), it is possible that the use of emoticons may follow different patterns in its use or as an amplifier of significance.
Another more recent international study based on a survey with a sample of Chinese subjects concluded that 88% of the respondents used emoticons (Qiao, 2010). These users preferred SMS over other channels and resources offered by F2F platforms to express emotions to others in their social networks. The findings also reveal that most of these users employed emoticons for comic purposes and as a substitute for non-verbal elements in face-to-face communication.
Cultural style in communication and use of emoticons
People learn to communicate their thoughts and feelings and the values of the culture they share (Gudykunst, 1997; Gudykunst et al, 2005). The continuum individualism-collectivism (Gudykunst et al, 1996; Hall, 1976; Hofstede & Bond, 1984; Hofstede et al, 2010) as a specific cultural marker has been considered by transcultural researchers the most important factor to explain the way in which culture can affect the communicative styles of individuals. These styles include the use of language, self-representation and non-verbal signs. Culture is expected to shape both verbal and non-verbal communication. Based on empirical studies, Gudykunst et al. (1996) claim that the styles used by dif- ferent individuals to communicate with each other vary depending on the cultures they come from (p.
511).
Researchers in transcultural and intercultural communication claim that cultures can be classified through the different ways in which their members communicate (Hall & Hall, 1990). Nordic cultures, which are characterized by individualism, such as the ones found in Germany, the Netherlands or Austria, where people tend to communicate with each other with a more direct approach, are consid- ered “low-context cultures” (LCC). Other cultures, with a more collectivist nature, such as those in Latin America and particularly in Asia —including China, Korea and Japan—, where native speakers tend to use a non-verbal style that includes body language and silence to communicate, are considered
“high-context cultures” (HCC). Similarly, Gudykunst et al. (1996) associate the communicative style of a “high cultural context” with the use of indirect, ambiguous, conservative and controlled language.
Conversely, the communicative style of a “low cultural context” tends to use direct, dramatic, open and feeling-dependent language. According to the definition of both categories, these authors agree on the fact that the cultures of China, Japan, Korea and other Asiatic cultures are more appropriately classified as cultures of “high context”, whereas countries like Germany, Switzerland or the United States are categorized as cultures of “low context”. Mediterranean countries such as France, Italy or Spain are halfway between LCC and HCC, according to transcultural researchers. The questions asked by transcultural scholars about the differences in communicative styles between cultures are interest- ing to us because they research the possible existence of cultural patterns in the preferential use of emoticons by people who come from different cultures.
Different cultures can shape different cognitive processes and practices in the individuals that are part of them because culture has an influence on experience. The results obtained from previous studies (Páez et al., 2004; Fernández et al., 2000) reveal that emotional reactions are perceived and referenced more intensely by people who belong to developed societies in which an individualistic culture is more appreciated and a feminine orientation is applied, with a social structure that favors low differences in power, etc. Also, different cultures reveal their emotions differently: facial expres- sions, postures and body movements, looks and physical contact play a significant role on social inter- action. These elements represent the visible forms in which cultural idiosyncrasy can be expressed.
On the other hand, Walther and D’Addario (2001: 344) concluded in their study that the real communicative effects of an emoticon are minimal within the context of the linguistic signs it might accompany. Later, Provine et al. (2007) found that emotional expression was subordinated to the pro- duction of spoken sentences. However, in contrast with this perspective, in other experimental studies Derks et al. (2008) observed the influence of emoticons as leading to more positive messages than in the cases in which they were not present. They concluded that emoticons could replace some of the functions of non-verbal behavior, complementing and improving the verbal message, although with- out the ability to contradict it.
Expression of emotions and reduction of framing effect
In a previous experimental study, Thompson and Foulger (1996) already examined the effect of emoticons on the perception of framing. The results of that study revealed that emoticons could in- deed alter the perception of framing and make the readers aware that they could interpret the message in a less serious tone. Therefore, their use represents a useful strategy to prevent cases of uninten- tioned framing. Ellensburg (2012) confirms with her findings that a very frequent use of emoticons in the professional field may reduce perceptions of power, competence, status and morality.
In sum, emoticons are basically graphic images with multiple and diverse uses which allow users to express their feelings through CMC. However, we still do not know with certainty what cultural or demographic aspects may affect their use in communication and under what emotional state users prefer to use certain emoticons to express the emotions they feel, such as anger, happiness or sadness.
3. G
ENERAL OBJECTIVES OF THE STUDYThe purpose of this study is to know and describe the use of emoticons in communication under the influence of the emotions embedded in the messages and the cultural background of the users. To do so, a questionnaire has been designed to collect information on whether emoticons are indeed used in SMS messages in CMC or SNS by users from different cultures; and to observe whether there are cultural patterns that determine the use of emoticons. Apart from considering the possible influence of variability in communicative cultural styles on the models of preferred use of emoticons, this study focuses particularly on the emotional state of users at the time of sending online messages. The analy-
sis of this research casts light on the question of whether the emotions felt by users may have an in- fluence on the choice of the type of emoticons that they consider more effective to convey their feel- ings.
Research questions
We are based on the arguments reviewed in previous studies which seem to suggest that, regard- less of cultural differences, humans tend to show a universal desire for maintaining optimism and a good mood in spite of crises and conflicts. Therefore, this study sets out the following questions for analysis and discussion:
Question 1: Do users in general, regardless of their cultural origin, who send positive messages through CMC choose to pair them with any type of non-verbal sign to convey and share their opti- mistic mood?
Question 2: Compared with the transmission of positive messages, do users in general reduce the frequency of use of sad non-verbal signs to convey negative messages through CMC to prevent an emphasis on the sad mood of those messages?
Anger is an instinctive feeling associated to moral balance that emerges when faced with an injus- tice and which causes a feeling of energy or impulsiveness in the individual. This leads the individual to act with intensity and immediately in order to actively solve the conflict situation. Considering this state of anger, we present:
Question 3:Do CMC users in general transmit messages which include verbal text directly rather than using emoticons with angry expressions?
In his original work, Hall (1976) argued that individuals from LCCs interpreted face-to-face communication as a way to approach problems and solve them. Conversely, people from HCCs transmitted messages without directly mentioning the problem verbally, and they preferred to use non-verbal expressions and let harmony flow. Considering these arguments, the following questions are presented:
Question 4: What are the preferences of Chinese individuals compared with Spanish individuals re- garding the use of emoticons to communicate messages about “anger” in CMC?
Question 5:What are the preferences of Chinese individuals compared with Spanish individuals re- garding the use of emoticons to communicate messages about “happiness” in CMC?
Question 6: What are the preferences of Chinese individuals compared with Spanish individuals re- garding the use of emoticons to communicate messages about “sadness” in CMC?
4. M
ETHODS The pre-testIn order to verify the correct identification of meanings conveyed through the emoticons most commonly used online, these characters were subject to a previous test for the unequivocal definition of the emotional message that each emoticon conveyed for CMC users. In the present study, the nine best-known and most widely used emoticons in CMC were used as a tool to examine the preferences of users when expressing their emotions in messages sent over the internet or social networks.
- Tool validation
The emoticons used for this study were, in the first place, created with ASCII symbols in the keyword to represent happiness :-), sadness :-( and anger ›:-( ; next, we used the faces known as smiley , frowny and angry ; and finally, we used the emojis with graphic facial elements that showed happiness , sadness and anger .
Table 1. Results of the pre-test for the identification of tools (N=40)
The reason why these elements were chosen is based on the research carried out by Rezabeck and Cochenour (1998), who claimed that, although the LISTSERV platform contained a higher variety of emoticons, these were the ones which appeared more frequently and, therefore, had a higher chance of creating a shared understanding of the meanings conveyed through them.
In this pilot test, 40 young individuals from a Spanish city with an approximate population of 200,000 people participated as volunteers. They were chosen randomly and their general profile can be summarized as follows: Average age=26.3 years (SD=7.26); 70% with a culturally Spanish origin and 30% from Asiatic cultures; 71.8% were students and 28.2% were workers; 60% were women and 40%
were men, with an average of 1-3 messages sent daily through CMC or SNS (Social networking ser- vices). The results obtained (Table 1) showed a high percentage of matches with regard to the identifi- cation of the meaning of the most commonly used emoticons, with a range of 79.5%-100%. This reveals the existence of a very high degree of agreement among the subjects of the pilot study on the different signs, the human emotions they represent and the meaning of their use in a communicative context.
This pilot study confirms the postulate by Rezabeck and Cochenour (1998) that claimed that the most widely and commonly used emoticons are more likely to produce a shared understanding of the meanings conveyed through them (Table 1).
Emoticons/emojis
Meaning identification match %
Happiness Anger Sadness Confusion
:-)
94.8 2.6 0 2.6100 0 0 0
100 0 0 0
›:-(
0 79.5 10.3 10.20 100 0 0
0 100 0 0
:-(
0 5.1 92.3 00 2.6 97.4 0
0 5.1 94.9 0
This verification of the agreed identification of the meaning of the emoticons proposed in the pi- lot study allowed us to use them later on in the instrumental design of the present study.
Survey study and instrumental design
A questionnaire was designed with 35 statements about common situations in the daily life of every human being. The objective was to know the preferences of users regarding the use of emoti- cons or emojis in the text messages that could best convey their mood when sending those messages through CMC or SNS. These statements were grouped as items into the following sections:
-Items with a communicative message about situations with the emotion of happiness In this part of the questionnaire, participants are asked about their preferences regarding the use of emoticons to express and convey their feelings through online messages in twelve situations in which they would probably feel happy, joyful, glad and optimistic. For example: “I have received a prize for a work well done” or “I have had a nice and unexpected surprise” or “Receiving love, kindness or a show of affec- tion”, etc. In order to indicate their response to each item in the twelve situations related to happiness, participants used a number scale with different happiness emoticons provided in the questionnaire.
Participants could choose different emoticons/emojis according to their order of preference to ex- press and convey their mood in messages through CMC or SNS: 1=“none”; that is, the message only uses verbal text; 2 = :-) ; 3 = ; 4 = .
- Items with a communicative message about situations with the emotion of sadness
In order to observe the preferences in the use of emoticons with which participants express and convey their feelings of sadness when sending messages through CMC or SNS, they are shown twelve sad situations in which humans feel affected, for example: “The death of a loved one”, or “Living in soli- tude” or “Ending a relationship”. The participants chose their order of preference regarding the use of emoticons/emojis included in the scale to express and convey their mood in messages through CMC or SNS for each item of these situations of sadness: 1=“none”, the message only uses verbal text; 2 = :-( ; 3 = ; 4 = .
- Items with a communicative message about situations with the emotion of anger
This section includes 6 items describing situations in which a person may experiment or react with anger, ire or a feeling of annoyance. For example: “I feel impotent against the actions of others that are incorrect, illegal or contrary to common sense in my opinion; but there is nothing I can do and I have to put up with them” or “I have been disrespected”. Participants were given a scale with different emoticons/emojis that they had to choose according to their preferences to respond to each of the six items related to the feeling of anger: 1=“none”, the message only uses verbal text; 2 = ›:-( 3 = ; 4 = .
- ICT use variable
The questionnaire included a section with three items about the use of social networks with a smartphone, use of the internet and information searchers on websites. It also included an item about the time spent online by participants whenever they went online, and an item that asked participants about their use of a foreign language to communicate with others through CMC and SNS.
-Demographic variables
Information was collected about participants through four demographic items: age, gender, cul- tural origins and working/studying situation.
Participants
In this survey the questionnaire was completed by 140 users of internet through CMC and SNS or SMS with mobile phones. The participants were volunteers and included users of online communi- cation, undergraduate university students and graduate students in Master’s Degree programs in an age range of 18-41 years, an average age of 23.14 years and SD=4.02. The gender distribution was 57.1%
women and 42.9% men. With regard to their cultural origins, 58.3% had been raised in a Spanish culture and 41.8% in a Chinese culture. Their professional profile was: graduate students=86.3%;
undergraduate students=2.1%; working and studying at the same time=11.7%. Participants were re- cruited in university libraries and classrooms in Spain.
5. RESULTS
Preliminary analysis
A preliminary analysis was performed with a contingency table to obtain the following descriptive results:
-Use of text, emoticons or emojis in angry messages
First of all, we analyzed the frequency with which respondents used non-verbal signs, such as the ASCII code sequence ›:-( on their keyboard, the angry emoticon or the angry emoji to send messages through CMC about daily situations that lead to anger which were presented in the ques- tionnaire.
In this analysis, with the exception of item 3, “Things are not going as well as I expected”, where the us- ers preferred the angry emoji (39.3%), the data reveal that respondents prefer the use of a direct verbal communicative style to convey their angry mood in three out of the six items, with percentages of 45%, 37.9% and 42.9% respectively, and with statistical significance between p < .001 and p < .01. In the multivariate test, the results were χ² (3)= 19.51; p < .001; Kendall’s W = .051 (Table 2).
Table 2. Preferred use of verbal text, signs or emoticons in angry messages through CMC (%)
Angry message item Text
only ›:-( χ² p
1) Similar negative things keep happening to me. 32.9 18.6 26.4 22.1 6.34 .096
2) I have been disrespected. 45.0 12.1 15.7 27.1 36.74 .000
3) Things are not going as well as I expected. 30.7 22.9 39.3 7.1 31.37 .000 4) Someone or something has interfered with or interrupted the
work or the task I was doing. 29.3 17.9 31.4 21.4 6.91 .075
5) I have to endure malaise, physical pain or mental stress. 37.9 22.9 23.6 15.7 14.45 .002 6) I feel impotent against the actions of others that are incorrect,
illegal or contrary to common sense in my opinion; but there is
nothing I can do and I have to put up with them. 42.9 17.1 21.4 18.6 24.34 .000
N =140, χ² (3) = 19.51; p < .001; Kendall’s W = .051
- Use of text, emoticons or emojis in happy messages
Table 3. Preferred use of verbal text, signs or emoticons in happy messages through CMC (%)
Happy message item Text only :-) χ² p
7) I have been given a prize for a job well done. 12.1 14.3 29.3 44.3 37.54 .000
8) The results were what I expected. 11.4 20.7 44.3 23.6 32.28 .000
9) I am praised with great esteem. 16.4 23.6 35.0 24.3 9.92 .019
10) I achieved what I had always wanted. 12.1 17.6 28.6 42.1 29.88 .000
11) I have had a nice and unexpected surprise. 12.1 16.4 30.7 40.0 27.99 .000 12) I have experienced a very pleasing physical and mental
sensation. 19.3 20.7 32.9 25.7 6.40 .093
13) I have been returned a favor for what I did. 27.9 27.9 32.9 10.7 15.90 .001 14) Receiving love, kindness or a show of affection. 12.1 17.9 32.9 37.1 23.82 .000
15) Meeting with friends or relatives. 29.3 10.7 37.9 22.1 22.17 .000
16) I have received some presents. 19.3 13.6 36.4 28.6 17.48 .000
17) Reality exceeds expectations. 18.6 20.7 25.0 35.0 9.0 .029
18) I have received good news. 13.6 17.1 37.9 31.4 22.34 .000
N=140; χ² (3) =43.97; p<.000; Kendall’s W =.11
This analysis reveals that respondents decide to use the smiley emoticon (32.9%-44.3%) and the happy emoji (35%-44.3%) respectively with a higher frequency to convey online the happiness they experience in positive situations (Table 3).
These results show statistical significance levels between p <.001 and p <.05, respectively, with mutlivariate analysis results of χ²(3) = 62.61; p<.001; Kendall’s W= .11.
- Use of non-verbal signs in sad messages
The analysis reveals that, unlike with the emotion of happiness, and instead of using non-verbal signs, respondents prefer to send CMC messages with verbal text about their feeling of sadness in most of the realistic situations presented in the questionnaire (34.5%-67.1%) (Table 4). These results show statistical significance levels ranging from p <.001 to p <.01. A multivariate analysis test shows the following results for our data: χ² (3)= 62.61; p <.001; Kendall’s W= .16 for the items of sad mes- sages.
Table 4. Preferred use of verbal text, signs or emoticons in sad messages through CMC (%)
Sad message item Text
only :-( χ² p
19) Things did not go well. The results was disappointing. 27.3 26.6 29.5 16.5 5.54 .136
20) The death of a loved one. 67.1 8.6 7.9 16.4 135.14 .000
21) Ending a relationship, parting ways. 61.4 17.9 7.9 12.9 101.88 .000
22) Realizing one’s own incapacity. 42.1 23.6 26.4 7.9 33.14 .000
23) Hearing or seeing something which fills me with sorrow and
pity. 32.9 24.3 26.4 16.4 7.71 .052
24) I have been reproached and rejected for something. 43.2 19.4 25.2 12.2 19.14 .000
25) I did not achieve what I wanted. 34.5 21.6 32.4 11.5 18.84 .000
26) Things are going worse than I anticipated. 36.2 24.6 26.1 13.0 14.92 .002 27) I feel bad because I hurt somebody through oversight. 38.1 19.4 24.5 18.0 14.06 .003
28) Serious illnesses in people close to me. 59.5 12.1 15.0 13.6 88.0 .000
29) Loneliness. 46.8 20.9 18.0 14.0 36.28 .000
30) The result of an unfair cause-effect relation 43.6 23.6 20.7 12.1 29.71 .000 N=140; χ² (3) =62.61; p<.000; Kendall’s W =.16
Transcultural studies about the use of emoticons and texts in online communication
In the following transcultural studies, the technique of analysis of variance (ANOVA) has been ap- plied to observe the way in which the cultural variable might affect the expression of emotions among Spanish and Chinese internet users when sending online messages through CMC.
- Internal homogeneity test of the items through approximation
Before performing the ANOVA study, the different dimensions were classified through a cluster analysis in order to reduce the amount of data. We grouped the items of the three primary emotions according to their approximation. That is, the items within each cluster were homogeneous and they showed significant differences when compared to the contents of other clusters (Vilà-Buenos et al., 2014). This classification made it possible to obtain the results of the different clusters based on the three basic universal emotions —happiness, anger, sadness. The dendrogram obtained with SPSS (Figure 1) shows that the items can be grouped into three emotion dimensions:
Cluster 1: Anger
3) Things are not going as well as I expected.
4) Someone or something has interfered with or interrupted the work or the task I was doing.
1) Similar negative things keep happening to me.
5) I have to endure malaise, physical pain or mental stress.
6) I feel impotent against the actions of others that are incorrect, illegal or contrary to common sense in my opinion; but there is nothing I can do and I have to put up with them.
2) I have been disrespected.
Figure 1. Cluster analysis
Cluster 2: Happiness
10) I achieved what I had always wanted.
11) I have had a nice and unexpected surprise.
8) The results were what I expected.
12) I have experienced a very pleasing physical and mental sensation.
14) Receiving love, kindness or a show of affection.
7) I have been given a prize for a job well done.
9) I am praised with great esteem.
13) I have been returned a favor for what I did.
16) I have received some presents.
18) I have received good news.
17) Reality exceeds expectations.
15) Meeting with friends or relatives.
Cluster 3: Sadness
20) The death of a loved one.
21) Ending a relationship, parting ways.
28) Serious illnesses in people close to me.
25) I did not achieve what I wanted.
26) Things are going worse than I anticipated.
19) Things did not go well. The results was disappointing.
22) Realizing one’s own incapacity.
30) The result of an unfair cause-effect relation
24) I have been reproached and rejected for something.
27) I feel bad because I hurt somebody through oversight.
23) Hearing or seeing something which fills me with sorrow and pity.
29) Loneliness.
ANOVA: Transcultural study of the use of texts, emoticons and emojis by Spanish vs Chinese users in messages conveying emotions through CMC
In order to develop this study, a transformation process was necessary to re-encode and recalcu- late the data through the creation of scales based on predefined measures.
The studies of analysis of variance were developed using the Spanish/Chinese culture as an inde- pendent variable and the emotional dimensions of “anger”, “happiness” and “sadness” as dependent variables, with four categories: “text only”; “keyboard-based emoticon using ASCII signs”, “emoti- con” and “facial emoji”.
“Anger” dimension
The results show that users in general prefer to use text only (M=.36; SD=.31) rather than the ASCII-based emoticon (M=.19; DT=.20), the “angry” emoticon (M=.27; DT=.22) or the facial emoji for anger (M=.19; DT=.20) to convey messages that involve angry moods that affect them. However, when comparing the preferences between different cultural groups, we have observed that Chinese respondents (m=.24; sd=.22) choose the ASCII sequence more often than Spanish users (m=.15;
sd=.18). This difference has a statistical significance of p < .01 (F[1.137]= 8.004; η²= .06). On the other hand, Spanish respondents (m= .24; sd= .22) chose the facial emoji more often than Chi- nese respondents (m= .12; sd= .16), with a significance of p< .001 (F[1.137]= 11.961; η²= .08) (Table 5-1).
“Happiness” dimension
Users as a whole express a preference for the use of the “smiley” emoticon (M= .34; SD= .22) to show happiness in online communicative messages, rather than using text only (M=.17; SD=.21), the ASCII sequence (M=.18; SD=.20) or the happy facial emoji (M=.30; SD=.24). In this dimen- sion of “happiness” we observe that Spanish respondents (m=.20; sd=.23) tend to use text only more often than Chinese participants (m=.13; sd=.18) with a significance close to p< 0.1 (Table 5-2).
“Sadness” dimension
In situations of sadness, users in our study generally preferred to use text only (M= .44; SD= .29) in their online communications. We have found that Spanish participants tend to convey their sadness through CMC with the use of text only (m= .53; sd= .27) more than Chinese respondents (m= .32;
sd= .27). The difference is significant, with p < .001 (F[1.137]= 20.185; η²= .13). On the other hand, Chinese respondents use the ASCII-based sequence (m= .27; sd= .19) more often than Spanish partic- ipants (m= .15; sd= .17), and this difference is also significant (p< .001; F[1.137]= 13.952; η²= .09).
Table 5: Results of ANOVA regarding the preferred use of text only, non-verbal signs, emoticons or emojis among Spanish and Chinese users to convey their emotions through CMC
1. “Anger” dimension Total Spanish Chinese F
(1.137) p η²
M SD m sd m sd
Angry text only .36 .31 .38 .30 .34 .32 .498 .481 .004
›:-(
.19 .20 .15 .18 .24 .22 8.004 .005 .06.27 .22 .24 .20 .30 .24 2.390 .124 .02
.19 .20 .24 .22 .12 .16 11.961 .001 .08
2. “Happiness” dimension
Happy text only .17 .21 .20 .23 .13 .18 2.877 .092 .021
:-)
.18 .20 .17 .22 .20 .17 .687 .409 .005.34 .22 .32 .22 .36 .21 1.330 .251 .01
.30 24 .31 .24 .29 .24 .188 .665 .001
3. “Sadness” dimension
Sad text only .44 .29 .53 .27 .32 .27 20.185 .000 .13
:-(
.20 .19 .15 .17 .27 .19 13.952 .000 .09.22 .17 .19 .17 .26 .18 5.589 .019 .04
.14 .17 .13 .17 .15 .18 .504 .479 .004
N= 139 n= 81 n= 58
At the same time, Chinese respondents also reveal that they prefer the use of the “frowny” emoti- con (m= 26; sd= .18) more than Spanish respondents (m= .19; sd= .17), and the difference was statistically significant (p< .05; F[1.137]= 5.589; η²= .09). However, the differences in the use of the sad facial emoji among Chinese participants (m= .15; sd= .18) and Spanish participants (m= .13;
sd= .17) was not statistically significant (Table 5-3).
6. DISCUSSION
The results obtained in our analysis provide a response to the research approach proposed prior to this study. It seems clear that, in spite of the crises and threats that this world lives through, hu- mans want to convey messages of hope to maintain an optimistic mood. The first research question is answered with the results of the analysis shown in Table 2, that reveal how CMC users who were surveyed in our study send, in general terms, the 12 types of messages about positive situations they may find themselves in. Most of the times, they do not only send positive verbal messages, but rein- force them with non-verbal signs like the smiley emoticon or a happy facial emoji. A simple smiling face can give a friendlier and nicer tone to a message. These results are consistent with those of other previous studies that showed that the use of emoticons has a high communicative power in social networks, because they can capture the attention faster than plain text and quickly convey positive feelings.
Compared with the communication of positive messages, Question 2 asks whether users in gen- eral reduce the frequency of use of non-verbal sad signs to communicate sad messages through CMC in order to prevent putting an emphasis on the sad emotion in those messages. Indeed, our study shows that CMC users generally reduce significantly the use of non-verbal signs, ASCII-based key- board emoticons (M=.20), the frowny emoticon (M=.22) and the sad facial emoji (M=.14) to convey sad messages. Under these circumstances, the subjects in our study prefer to use direct verbal texts (M=.44).
Similarly, our study confirms the hypothesis presented in Question 3. The results of the analysis show that CMC users generally choose to send angry messages with verbal text (M=.36) in which they can express themselves directly, rather than using non-verbal signs like the ASCII-based keyboard emoticon (M=.19), the angry emoticon (M=.26) or the angry facial emoji (M=.18).
Also, Question 4 has been answered with the data that reveal that in general terms, and without differences between the different subgroups, users prefer the use of verbal texts (M=.36) to convey angry messages. However, when comparing the most commonly used non-verbal signs in the two subgroups of participants, we have observed that Chinese respondents choose the ASCII sign ›:-(
more often (m=.24) than their Spanish counterparts (m=.15). Conversely, Spanish participants opt for the angry facial emoji ” (m=.24) more often than the Chinese (m=.12). These data were statistical- ly significant with p< .01 and p< .001, respectively.
With regard to Question 5, about whether the cultural background may affect the way in which CMC users transmit online messages about a happy feeling, the results of the analysis show that both cultural groups of participants like the smiley emoticon and the happy facial emoji without distinction.
This confirms the universal nature of online communication of good feelings about positive situa- tions.
Finally, our findings can cast light on the cultural aspect discussed in Question 6 about the differ- ences between users from Spain and from China. The results show that, in general terms, both groups use texts as a first option, although Spanish respondents use plain text only (m=.53) significantly more than Chinese respondents (m=.32). However, with regard to the non-verbal use of the ASCII-based symbol :-( and the frowny emoticon or the sad facial emoji to convey sad messages online, Chinese participants used these resources more often than their Spanish counterparts.
In sum, the results of this study, which is based on premises from previous studies, show that un- like the universal attribution of the emotion of happiness to the positive situations experienced by different individuals, there are clear cultural differences associated with the communication of sadness.
In this regard, the communicative style observed among the Spanish respondents to our questionnaire is oriented towards the low-context cultural end —in the continuum of cultural variability— com- pared with the communicative style of the Chinese culture, whose cultural prototype is placed on the opposite extreme (high context), in which there is a prevalence of an indirect communication style.
The previous observation is particularly clear in this study when we approach the communication of sadness through verbal vs non-verbal signs in CMC messages. Since the high-context communica- tive style in a collectivist culture (Hall, 1976) provides more information related to the physical envi- ronment, individuals in HCCs, like the Chinese, put their trust in non-verbal communication rather than in verbal communication, whereas a low context communicative style, characteristic of individu- alistic cultures, is identified with the opposite characteristics; that is, the meaning of messages is inter- preted through what is written or said in verbal communication.
The same thing happens with face-to-face communication: HCCs tend to use non-verbal strate- gies to convey the meaning that may be transmitted through gestures, body language, silence or sym- bolic actions. This study has observed that people in the Chinese culture, which is more collectivist, generally use emoticons, rather than facial emojis —as we anticipated—, with more frequency and intensity than Spanish participants in the exchange of messages through CMC and SNS. On the other hand, in users with a cultural tendency relatively more individualistic, like Spanish culture when com- pared with Chinese culture, their communicative style would also follow the lines of a LCC.
Finally, with regard to a more common use of the angry facial emoji by Spanish respondents, we can interpret that the extroverted expression of that emoji is part of a direct communicative style, which characterizes LCCs, in an attempt to intensify the angry feeling more directly and unequivocally, rather than using a tenuous insinuation of the feelings that might be conveyed with other emoticons that use iconographic symbols.
7. CONCLUSIONS
This study has revealed data about some aspects of the use of non-verbal signs, like emoticons and emojis, among users of CMC and SNS to communicate messages about states of mind connected to different universal primary emotions: happiness, sadness and anger. The theory of cultural variabil- ity included here as a theoretical framework has been used as a basis to research whether there are different models for the preferential use of these non-verbal signs vs the option of using only plain text among users with different cultural backgrounds during the process of interaction through mes- sages on CMC and/or SNS to convey the emotions they feel in different situations.
The technique of cluster analysis has been used as part of the methodological process. It revealed the internal consistence of the items with the number scale of the questionnaire. Thanks to this tech- nique, the items are classified according to their approximation with each other to establish three emotional dimensions —happiness, sadness and anger—, which are considered by social psychologists the three primary human emotions shared by everyone, regardless of their cultural origin. Unlike these universally-shared primary emotions, there are others, like guilt, disgust, shame, etc., which according to the experts are culture-specific and must be analyzed within each cultural context.
Confirming the existence of the three clearly defined and different dimensions has allowed the re- search analyst of this study to perform an ANOVA test to examine the existence of cultural patterns in the use of non-verbal signs, such as emoticons, among users with different cultural backgrounds, to convey and exchange personal messages related to their emotions.
The findings are far from definitive due, on the one hand, to the small number of participants (N=140) who collaborated by completing the questionnaire, and on the other hand, to the necessarily reduced range of emoticons and emojis included in the instrumental design of this research. Also, the use of Spanish rather than Chinese may cause some variations in the interpretation of the exact mean- ing of the items by Chinese participants, although they were students enrolled in graduate and under- graduate programs in Spanish universities. These limitations aside, the data shown in the analysis give us an idea of the type of emotional context preferred by CMC and/or SNS users from different cul- tures and what type of emoticon/expressive emoji vs plain text they prefer in online messages to convey their mood.
8. R
EFERENCESAdrianson, L. (2001). Gender and computer-mediated communication: Group processes in problem, Computers in Human Behavior, 17(1):71-94.
Amaghlobeli, N. (2012). Linguistic features of typographic emoticons in SMS discourse, Theory and Practice in Language Studies, 2 (2), pp. 348-354.
Constantin, C., Kalyanaraman, S., Stavrositu, C. & Wagoner, N. (2002). To be or not to be emotional:
Impression formation effects of emoticons in moderated chatrooms. Paper presented at the Communication Technology and Policy Division at the 85th Annual Convention of the Association for Educa- tion in Journalism and Mass Communication, Miami Beach, Florida, in 2002 August (Fecha consulta el 08.11. 2015).
Crystle, D. (2001). Language and the Internet. Cambridge University Press, Cambridge.
Derks, D., Fischer, A. H. & Bos A. E. R. (2007). The role of emotion in computer-mediated commu- nication: A review. Computers in Human Behavior, doi:10.1016/j.chb.2007.04.004. Last accessed 09. 12. 2015.
Dresner, E. & Herring, S. C. (2010), Functions of the nonverbal in CMC: Emoticons and illocutionary Force. Communication Theory, 20, pp. 249-268. doi: 10.1111/j.1468-2885.2010.01362.x Last ac- cessed 19. 12. 2015.
Duggan, B. U. M. & Smith, A. (2013). Social media update 2013. Pew Research Internet Project, 7-9.
Ekman, P. (1993). Facial expression and emotion. American Psychologist, 48(4), pp. 384-392.
Ekman, P. & Oster H. (1997). Expresiones faciales de la emoción, Annual Review of Psychology, 30, pp.
527-554.
Ellensburg, R. (2012). Emoticons as social information: The effect of emoticons on interpersonal perception in computer-mediated business communication. TFM de programa de investigación en psicología, Universidad de Amsterdam.
http://dare.uva.nl/cgi/arno/show.cgi?fid=463478 Last accessed 06. 02. 2016.
Fahlman, S. E. (1982). Smiley Lore :-) (unpublished). https://www.cs.cmu.edu/~sef/sefSmiley.htm Last accessed 10. 06. 2015.
Fernández, I., Zubieta, E. & Páez, D. (2000). Expresión e inhibición emocional en diferentes culturas.
En D. Páez & M. M. Casullo (Comps.), Cultura y alexitimia: ¿Cómo expresamos aquello que sentimos?
(pp. 73-98). Buenos Aires: Paidós.
Ferris, S. (1996). Women on-line: Cultural and relational aspects of women’s communication in on- line discussion. Interpersonal Computing and Technology, vol. 4(3-4), pp. 29-40.
http://www.helsinki.fi/science/optek/1996/n3/ferris.txt Last accessed 08. 08. 2015.
Gudykunst, W. B. (1997). Cultural variability in communication: An introduction. Communication Re- search, 2(4), pp. 327-348.
Gudykunst, W. B., Lee, C. M., Nishida, T. & Ogawa, N. (2005). Theorizing about intercultural communica- tion. Thousand Oaks, CA: Sage.
Gudykunst, W. B., Matsumoto, Y., Ting-Toomey, S., Nishida, T., Kim, K. & Heyman, S. (1996). The influence of cultural individualism-collectivism, self construals, and individual values on com- munication styles across cultures. Human Communication Research, 22(4), pp. 510-543.
Hall, E. T. (1976). Beyond culture. Garden City, NY: Doubleday.
Hall, E. T. & Hall, M. R. (1990). Understanding cultural differences. Germans French and Americans (Vol. 54).
Dispoinble en http://doi.org/10.5860/CHOICE.28-2221. Last accessed 14. 09. 2008.
Hofstede, G. & Bond, M. H. (1984). Hofstede’s culture dimensions. Journal of Cross-Cultural Psychology, 15(4), pp. 417-433.
Hofstede, G., Hofstede, G. J. & Minkov, M. (2010). Culture and organizations: Soft ware of the mind. Inter- national cooperation and its importance for survival. NY: McGraw-Hill Companies.
Huffaker, D. A. & Calvert, S. L. (2006). Gender, identity and language use in teenage blogs. Journal of Computer Mediated Communication, 10(2), DOI: 10.1111/j.1083-6101.2005.tb00238 Last accessed 20.
08. 2017.
Jenson, S. (2005). Default thinking: Why consumer products fail. In R. Harper, A. Taylor, & L. Palen (Eds.), The inside text: Social perspectives on SMS in the mobile age. pp 305-325 London: Kluwer.
Jibril, T. A. & Abdullah, M. H. (2013). Relevance of emoticons in computer mediated communication contexts: An overview. Asian Social Science, 9(4), pp. 201-207.
Kasensniemi, E. L. & Rautiainen, P. (2002). Mobile culture of children and teenagers in Finland. N. Y.: Cam- bridge University Press.
Lee, C. (2003). How does instant messaging affect interaction between the genders? Stanford, CA: The Mercury Project for Instant Messaging Studies at Stanford University.
http://www.stanford.edu/class/pwr3-25/group2/projects/lee.html Last accessed 08. 12. 2015.
Ling, R. (2005). Mobile communication vis-à-vis teen emancipation, peer group integration and devi- ance. In R. Harper, A. Taylor, & L. Palen (Eds.), The inside text: Social perspectives on SMS in the mo- bile age (pp. 175–189). London: Kluwer.
Lo, S. (2008). The nonverbal communication functions of emoticons in computer-mediated commu- nication. Cyber Psychology & Behavior, 11 (5), pp. 595-597.
Páez, D., Fernández, I., Ubillos, S. & Zubieta, E.(eds.) (2004): Psicología social, cultura y educación. Madrid:
Pearson Educación.
Provine, R. R., Spencer, R. J. & Mandell, D. L. (2007). Emotional expression online: Emoticons punc- tuate website text messages, Journal of Language and Social Psychology, 26, pp. 299-307.
Ptaszynski, M., MacIejewski, J., Dybala, P., Rzepka, R. & Araki, K. (2010). CAO: A fully automatic emoticon analysis system based on theory of kinesics. IEEE Transactions on Affective Computing, 1(1), pp. 46-59.
Qiao, Q. (2010). The adoption and use of SMS among Chinese teenagers. Memoire: Universite de Montre- al. https://papyrus.bib.umontreal.ca/xmlui/handle/1866/3554 Last accessed 06. 02. 2016.
Rezabek, L. L. & Cochenour, J. J. (1998). Visual cues in computer-mediated communication: Supple- menting text with emoticons. Journal of Visual Literacy, 18, pp. 201-215.
Thompson, P. A. & Foulger, D. A. (1996). Effects of pictographs and quoting on flaming in electronic mail. Computer in Human Behavior, 12, pp. 225-243.
Tossell, Ch. C., Kortum, P., Shepherd, C., Barg-Walkow, L. H., Rahmati, A. & Zhong, L. (2012). A longitudinal study of emoticon use in text messaging from smartphones, Computer in Human Be- havior, 28, pp. 659-663.
Vilà-Baños, R., Rubio-Hurtado, M.J., Berlanga.Silvente, V. & Torrado-Fonseca, M. (2014). Cómo aplicar un cluster jerárquico en SPSS, REIRR, 7(1), pp 113-127.
Walter, J. B. & D’Addario, K. P. (2001). The impacts of emoticons on message interpretation in com- puter-mediated communication. Social Science Computer Review.
http://doi.org/10.1177/089443930101900307 Last accessed 15. 04. 2015.